Perplexity AI has released pplx-decider-v1.3, an updated multimodal classification model built on Qwen3 8.27B. This new version offers enhanced performance and robustness for decision-making tasks, and introduces a 'Locate' capability to extract verbatim text spans with character offsets and confidence scores. The model requires Python 3.12+, a CUDA GPU with significant memory, and a Hugging Face token for access. AI
IMPACT Enhances classification and extraction capabilities for multimodal tasks, potentially improving customer support and data analysis workflows.
RANK_REASON This is a model release from a company that is not a frontier lab, and it is presented as a tool for classification and extraction.
Read on Hugging Face Trending Models →
- CUDA
- Hugging Face
- huggingface_hub
- perplexity-ai/pplx-decider-v1.3
- Pillow
- Python
- PyTorch
- Qwen3 8.27B
- safetensors
- torchvision
- transformers
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →